scholarly journals Finding Meanings in Low Dimensional Structures: Stochastic Neighbor Embedding Applied to the Analysis of Indri indri Vocal Repertoire

Animals ◽  
2019 ◽  
Vol 9 (5) ◽  
pp. 243
Author(s):  
Daria Valente ◽  
Chiara De Gregorio ◽  
Valeria Torti ◽  
Longondraza Miaretsoa ◽  
Olivier Friard ◽  
...  
Keyword(s):  
Cluster Analysis ◽  
A Priori ◽  
Computational Method ◽  
Vocal Repertoire ◽  
Acoustic Features ◽  
Acoustic Structure ◽  
Spectral Coefficients ◽  
Indri Indri ◽  
Hard Clustering ◽  
Low Dimensional

Although there is a growing number of researches focusing on acoustic communication, the lack of shared analytic approaches leads to inconsistency among studies. Here, we introduced a computational method used to examine 3360 calls recorded from wild indris (Indri indri) from 2005–2018. We split each sound into ten portions of equal length and, from each portion we extracted spectral coefficients, considering frequency values up to 15,000 Hz. We submitted the set of acoustic features first to a t-distributed stochastic neighbor embedding algorithm, then to a hard-clustering procedure using a k-means algorithm. The t-distributed stochastic neighbor embedding (t-SNE) mapping indicated the presence of eight different groups, consistent with the acoustic structure of the a priori identification of calls, while the cluster analysis revealed that an overlay between distinct call types might exist. Our results indicated that the t-distributed stochastic neighbor embedding (t-SNE), successfully been employed in several studies, showed a good performance also in the analysis of indris’ repertoire and may open new perspectives towards the achievement of shared methodical techniques for the comparison of animal vocal repertoires.

scholarly journals Computational Method for Wavefront Sensing Based on Transport-Of-Intensity Equation

Photonics ◽  
2021 ◽  
Vol 8 (6) ◽  
pp. 177
Author(s):  
Iliya Gritsenko ◽  
Michael Kovalev ◽  
George Krasin ◽  
Matvey Konoplyov ◽  
Nikita Stsepuro
Keyword(s):  
A Priori ◽  
Spherical Wave ◽  
Computational Method ◽  
Optical Systems ◽  
Radius Of Curvature ◽  
Imaging Method ◽  
Wavefront Sensing ◽  
Phase Imaging ◽  
Transport Of Intensity Equation ◽  
Priori Information

Recently the transport-of-intensity equation as a phase imaging method turned out as an effective microscopy method that does not require the use of high-resolution optical systems and a priori information about the object. In this paper we propose a mathematical model that adapts the transport-of-intensity equation for the purpose of wavefront sensing of the given light wave. The analysis of the influence of the longitudinal displacement z and the step between intensity distributions measurements on the error in determining the wavefront radius of curvature of a spherical wave is carried out. The proposed method is compared with the traditional Shack–Hartmann method and the method based on computer-generated Fourier holograms. Numerical simulation showed that the proposed method allows measurement of the wavefront radius of curvature with radius of 40 mm and with accuracy of ~200 μm.

Convergence of management practices in India’s IT sector with ana priorivalidation

2018 ◽  
Vol 12 (4) ◽  
pp. 402-421
Author(s):  
Jayashree Mahesh ◽  
Anil K. Bhat
Keyword(s):  
Cluster Analysis ◽  
Operations Management ◽  
Management Practices ◽  
Qualitative Interviews ◽  
A Priori ◽  
Statistical Testing ◽  
Content Type ◽  
Empirical Survey ◽  
Different Types ◽  
Group A

PurposeThe purpose of this paper is to document similarities and differences between management practices of different types of organizations in India’s IT sector through an empirical survey. The authors expected these differences to be significant enough for us to be able to groupa priorithis set of companies meaningfully through cluster analysis on the basis of the similarity of their management practices alone.Design/methodology/approachUsing a mixed-methods approach, 73 senior-level executives of companies working in India’s IT sector were approached with a pretested questionnaire to find out differences on eighteen management practices in the areas of operations management, monitoring management, targets management and talent management. The different types of organizations surveyed were small and amp; medium global multinationals, large global multinationals, small and medium Indian multinationals, large Indian multinationals and small and medium local Indian companies. The differences and similarities found through statistical testing were further validateda priorithrough cluster analysis and qualitative interviews with senior-level executives.FindingsThe management practices of multinationals in India are moving toward Western management practices, indicating that management practices converge as the organizations grow in size. Though the practices of large Indian multinationals were not significantly different from those of global multinationals, the surprising finding was that large Indian multinationals scored better than global multinationals on a few practices. The practices of small and medium Indian companies differed significantly from those of other types of organizations and hence they formed a cluster.Practical implicationsThe finding that large Indian IT multinationals have an edge over global multinationals in certain people management practices is a confirmation of the role of human resource practices in their current success and their continuing competitive advantage.Originality/valueThis is perhaps the first study of its kind to document state of specific management practices across different types of organizations in India’s IT sector and then use measures on these practices to group a priori these organizations for validation.

scholarly journals ABOUT PARAMETRIZATION OF SELECTION OF SIGNIFICANT CLUSTERS

2019 ◽  
Vol 4 (1) ◽  
pp. 64-67
Author(s):  
Pavel Kim
Keyword(s):  
Cluster Analysis ◽  
A Priori ◽  
Multidimensional Data ◽  
Number Of Clusters ◽  
Clustering Model ◽  
Measure Of Similarity ◽  
Selection Of

One of the fundamental tasks of cluster analysis is the partitioning of multidimensional data samples into groups of clusters – objects, which are closed in the sense of some given measure of similarity. In a some of problems, the number of clusters is set a priori, but more often it is required to determine them in the course of solving clustering. With a large number of clusters, especially if the data is “noisy,” the task becomes difficult for analyzing by experts, so it is artificially reduces the number of consideration clusters. The formal means of merging the “neighboring” clusters are considered, creating the basis for parameterizing the number of significant clusters in the “natural” clustering model [1].

scholarly journals Thermoelectric Properties of InA Nanowires from Full-Band Atomistic Simulations

Molecules ◽  
2020 ◽  
Vol 25 (22) ◽  
pp. 5350
Author(s):  
Damiano Archetti ◽  
Neophytos Neophytou
Keyword(s):  
Surface Roughness ◽  
Thermoelectric Power ◽  
Power Factor ◽  
Large Scale ◽  
Large Diameter ◽  
Computational Method ◽  
Thermoelectric Power Factor ◽  
Full Band ◽  
Low Dimensional ◽  
Power Factor Improvement

In this work we theoretically explore the effect of dimensionality on the thermoelectric power factor of indium arsenide (InA) nanowires by coupling atomistic tight-binding calculations to the Linearized Boltzmann transport formalism. We consider nanowires with diameters from 40 nm (bulk-like) down to 3 nm close to one-dimensional (1D), which allows for the proper exploration of the power factor within a unified large-scale atomistic description across a large diameter range. We find that as the diameter of the nanowires is reduced below d < 10 nm, the Seebeck coefficient increases substantially, as a consequence of strong subband quantization. Under phonon-limited scattering conditions, a considerable improvement of ~6× in the power factor is observed around d = 10 nm. The introduction of surface roughness scattering in the calculation reduces this power factor improvement to ~2×. As the diameter is decreased to d = 3 nm, the power factor is diminished. Our results show that, although low effective mass materials such as InAs can reach low-dimensional behavior at larger diameters and demonstrate significant thermoelectric power factor improvements, surface roughness is also stronger at larger diameters, which takes most of the anticipated power factor advantages away. However, the power factor improvement that can be observed around d = 10 nm could prove to be beneficial as both the Lorenz number and the phonon thermal conductivity are reduced at that diameter. Thus, this work, by using large-scale full-band simulations that span the corresponding length scales, clarifies properly the reasons behind power factor improvements (or degradations) in low-dimensional materials. The elaborate computational method presented can serve as a platform to develop similar schemes for two-dimensional (2D) and three-dimensional (3D) material electronic structures.

scholarly journals NEDD: a network embedding based method for predicting drug-disease associations

2020 ◽  
Vol 21 (S13) ◽  
Author(s):  
Renyi Zhou ◽  
Zhangli Lu ◽  
Huimin Luo ◽  
Ju Xiang ◽  
Min Zeng ◽  
...  
Keyword(s):  
Drug Repositioning ◽  
Computational Method ◽  
Heterogeneous Information ◽  
Gold Standard Dataset ◽  
Disease Similarity ◽  
Disease Associations ◽  
Meta Path ◽  
Approved Drugs ◽  
Low Dimensional ◽  
Novel Associations

Abstract Background Drug discovery is known for the large amount of money and time it consumes and the high risk it takes. Drug repositioning has, therefore, become a popular approach to save time and cost by finding novel indications for approved drugs. In order to distinguish these novel indications accurately in a great many of latent associations between drugs and diseases, it is necessary to exploit abundant heterogeneous information about drugs and diseases. Results In this article, we propose a meta-path-based computational method called NEDD to predict novel associations between drugs and diseases using heterogeneous information. First, we construct a heterogeneous network as an undirected graph by integrating drug-drug similarity, disease-disease similarity, and known drug-disease associations. NEDD uses meta paths of different lengths to explicitly capture the indirect relationships, or high order proximity, within drugs and diseases, by which the low dimensional representation vectors of drugs and diseases are obtained. NEDD then uses a random forest classifier to predict novel associations between drugs and diseases. Conclusions The experiments on a gold standard dataset which contains 1933 validated drug–disease associations show that NEDD produces superior prediction results compared with the state-of-the-art approaches.

A SINGULAR PROBLEM IN INCOMPRESSIBLE NONLINEAR ELASTOSTATICS

2000 ◽  
Vol 10 (08) ◽  
pp. 1181-1207 ◽  
Author(s):  
ADAIR R. AGUIAR ◽  
ROGER L. FOSDICK
Keyword(s):  
Rectangular Cross Section ◽  
A Priori ◽  
Small Neighborhood ◽  
Computational Method ◽  
Singular Problem ◽  
Lateral Side ◽  
Local Shape ◽  
Incompressible Elasticity ◽  
Rigid Surfaces ◽  
Selective Integration

This paper represents a contribution to the numerical treatment of problems in incompressible elasticity theory for large deformations. We are especially concerned about the solution of plane problems with corners. A review of the literature on these problems indicates that the behavior of the solution in the vicinity of a corner is given little attention. We investigate the solution of the compressed bonded block problem corresponding to the compression of an incompressible elastic block of rectangular cross-section and infinite transverse length between two opposing bonded rigid surfaces, with the two remaining lateral faces traction-free. We are especially interested in the behavior at a corner where a bonded end is adjacent to a free lateral side. We employ a finite element method based on a reduced and selective integration technique with penalization to construct a numerical solution for this problem. Our computational method converges everywhere except in a small neighborhood of the corner. We appeal to an elementary a priori inequality concerning the angle of shear to show that the numerical calculations in this neighborhood are inaccurate and need a more refined study. Based on the inequality, we offer a conjecture concerning the local shape of the deformed free lateral surface at the corner.

Comparing dietary patterns derived by two methods and their associations with obesity in Polish girls aged 13–21 years: the cross-sectional GEBaHealth study

2016 ◽  
Vol 137 (3) ◽  
pp. 182-189 ◽  
Author(s):  
Lidia Wadolowska ◽  
Joanna Kowalkowska ◽  
Jolanta Czarnocinska ◽  
Marzena Jezewska-Zychowicz ◽  
Ewa Babicz-Zielinska
Keyword(s):  
Cluster Analysis ◽  
Dietary Patterns ◽  
Central Obesity ◽  
Reference Group ◽  
A Priori ◽  
Multiple Logistic Regression Analysis ◽  
Total Sample ◽  
Data Driven ◽  
Low Fat ◽  
Cross Sectional

Aims: To compare dietary patterns (DPs) derived by two methods and their assessment as a factor of obesity in girls aged 13–21 years. Methods: Data from a cross-sectional study conducted among the representative sample of Polish females ( n = 1,107) aged 13–21 years were used. Subjects were randomly selected. Dietary information was collected using three short-validated food frequency questionnaires (FFQs) regarding fibre intake, fat intake and overall food intake variety. DPs were identified by two methods: a priori approach (a priori DPs) and cluster analysis (data-driven DPs). The association between obesity and DPs and three single dietary characteristics was examined using multiple logistic regression analysis. Results: Four data-driven DPs were obtained: ‘Low-fat-Low-fibre-Low-varied’ (21.2%), ‘Low-fibre’ (29.1%), ‘Low-fat’ (25.0%) and ‘High-fat-Varied’ (24.7%). Three a priori DPs were pre-defined: ‘Non-healthy’ (16.6%), ‘Neither-pro-healthy-nor-non-healthy’ (79.1%) and ‘Pro-healthy’ (4.3%). Girls with ‘Low-fibre’ DP were less likely to have central obesity (adjusted odds ratio (OR) = 0.36; 95% confidence interval (CI): 0.17, 0.75) than girls with ‘Low-fat-Low-fibre-Low-varied’ DP (reference group, OR = 1.00). No significant associations were found between a priori DPs and overweight including obesity or central obesity. The majority of girls with ‘Non-healthy’ DP were also classified as ‘Low-fibre’ DP in the total sample, in girls with overweight including obesity and in girls with central obesity (81.7%, 80.6% and 87.3%, respectively), while most girls with ‘Pro-healthy’ DP were classified as ‘Low-fat’ DP (67.8%, 87.6% and 52.1%, respectively). Conclusion: We found that the a priori approach as well as cluster analysis can be used to derive opposite health-oriented DPs in Polish females. Both methods have provided disappointing outcomes in explaining the association between obesity and DPs. The cluster analysis, in comparison with the a priori approach, was more useful for finding any relationship between DPs and central obesity. Our study highlighted the importance of method used to derive DPs in exploring associations between diet and obesity.

scholarly journals A Commentary on Noyce, Küssner and Sollich: Quantifying Shapes

2013 ◽  
Vol 8 (2) ◽  
pp. 158
Author(s):  
Christian Hennig
Keyword(s):  
Cluster Analysis ◽  
Low Dimensional

This review focuses on the statistical aspects of "Quantifying Shapes". A number of decisions made by the authors are discussed regarding the choice of methods and how exactly they are applied and their results interpreted. Most space is devoted to issues of cluster analysis and low dimensional embedding, with further remarks concerning GP regression and classification.

scholarly journals A cluster-genetic programming approach for detecting pulmonary tuberculosis

2021 ◽  
Vol 14 (1) ◽  
pp. 71-88
Author(s):  
Adane Nega Tarekegn ◽  
Tamir Anteneh Alemu ◽  
Alemu Kumlachew Tegegne
Keyword(s):  
Machine Learning ◽  
Cluster Analysis ◽  
Genetic Programming ◽  
A Priori ◽  
Real Data ◽  
Health Concern ◽  
Programming Approach ◽  
Data Set ◽  
Middle Income ◽  
Gp Model

Tuberculosis (TB) remains a global health concern. It commonly spreads through the air and attacks low immune bodies. TB is the most common and known health problem in low and middle-income countries. Genetic programming (GP) is a machine learning model for discovering useful relationships among the variables in complex clinical data. It is more appropriate in a circumstance when the form of the solution model is unknown a priori. The main objective of this study is to develop a model that can detect positive cases of TB suspected patients using genetic programming approach. In this paper, Genetic Programming (GP) is exploited to identify the presence of positive cases of tuberculosis from the real data set of TB suspects and hospitalized patients. First, the dataset is pre-processed, and target variables are identified using cluster analysis. This data-driven cluster analysis identifies two distinct clusters of patients, representing TB positive and TB negative. Then, GP is trained using the training datasets to construct a prediction model and tested with a separate new dataset. With the 30 runs, the median performance of GP on test data was good (sensitivity=0.78, specificity=0.95, accuracy=0.89, AUC=0.91). We find that GP shows better performance in predicting TB compared to other machine learning models. The study demonstrates that the GP model might be used to support clinicians to screen TB patients.

Sign in / Sign up

Export Citation Format

Share Document